- Email: [email protected]
Current challenges in the diagnosis of autoimmune pancreatitis
PANCREATIC PATHOLOGY
immunosuppressive therapy.3e5 Nevertheless, in some cases the disease shows overlapping features with pancreatic cancer, and it is in these cases that histopathology plays a critical role. Additionally, the lack of a specific biomarker for type 2 autoimmune pancreatitis (see below) has meant that these patients are frequently subjected to a biopsy, and occasionally, a pancreatic resection.
Current challenges in the diagnosis of autoimmune pancreatitis Rajib Gupta
Variants of autoimmune pancreatitis
Vikram Deshpande
Two distinct variants of autoimmune pancreatitis are recognized, labeled type 1 and type 2.5e10 The two variants are clinically, serologically, as well as histologically distinctive, although both forms of the disease respond to immunosuppressive therapy. In fact, based on these differences, it has been suggested that the two “variants” represent distinct diseases, a position that we endorse. The concept of type 1 autoimmune pancreatitis was originally proposed in Japan, while type 2 autoimmune pancreatitis was championed in Europe, the latter primarily on the basis of histopathologic findings. Notably, type 2 autoimmune pancreatitis is uncommon in the Asian population.
Abstract Autoimmune pancreatitis (AIP), an inflammatory disease of the pancreas first described in Japan, is characterized by a tumefactive lesion in the pancreas, thus clinically resembling pancreatic cancer. Currently, it is classified into two variants e type 1 AIP and type 2 AIP, each with distinct clinico-pathologic and epidemiologic features. Type 1 AIP is a part of IgG4 related disease, a generalized multi-systemic disease, and recapitulates its clinical and pathologic features. Type 2 AIP is an isolated pancreas-centric disease, and is not associated with elevated tissue or serum IgG4. Histologically, type 1 AIP is characterized by storiform-type fibrosis and obliterative phlebitis, while type 2 disease shows granulocytic epithelial lesions. Even when AIP is suspected, a diagnosis should seldom be made in isolation, but should be rendered only after collating clinical and radiologic data. Steroids remain the mainstay of treatment for both variants of AIP, although other immunosuppressive drugs like rituximab have proven successful. In this review, we summarize the current knowledge of this entity and discuss diagnostic challenges, with particular emphasis on the interpretation of needle biopsies. Recognition of AIP as a mimic of pancreatic cancer is imperative to reduce unnecessary morbidity and mortality associated with pancreatic surgery.
Type 1 autoimmune pancreatitis This is the pancreatic manifestation of IgG4 related disease. IgG4 related disease is a multi-systemic disease characterized by the presence of multifocal fibroinflammatory mass lesion(s) that are responsive to immunosuppressive therapy.11,12 Type 1 autoimmune pancreatitis bears all the clinical, serologic, and histologic hallmarks of IgG4 related disease. This is a disease of elderly men and the majority of these patients (60e75%) present with painless obstructive jaundice. A small proportion present with acute pancreatitis. Rarely the patient is entirely asymptomatic, the disease detected on cross-sectional imaging.
Keywords autoimmune pancreatitis; IgG4; IgG4 related disease
Histopathology Virtually all pancreatic resections from individuals with type 1 autoimmune pancreatitis reveal a dense lymphoplasmacytic infiltrate, storiform-type fibrosis and obliterative phlebitis (Table 1).8,9,13,14 In the authors’ experience, only patients pre-treated with immunosuppressive therapy lack a brisk inflammatory infiltrate. The infiltrate is generally diffusely distributed throughout the pancreas, although accentuation in the periductal region is often appreciable (Figures 1e3). Interestingly, although significant inflammation surrounds pancreatic ducts, the ductal epithelium is well preserved. An increased number of eosinophils are invariably found, and in rare instances eosinophils may dominate the histologic picture, and in numbers sufficient to raise the possibility of an allergic or an unrelated connective tissue disease. The disease is characterized by dense fibrosis, generally laid down in a storiform pattern (Figure 1). This is a pattern familiar to pathologists, most notably seen in dermatofibrosarcoma protuberans. The pattern is likely fashioned by the deposition of fibroblasts or myofibroblasts, the latter representing a consistent finding in autoimmune pancreatitis. Hyalinizing or keloidal type fibrosis is uncommon in this variant of autoimmune pancreatitis. Perhaps the most unique feature of autoimmune pancreatitis is obliterative phlebitis. Partial obliteration (>obliteration of 50%
Introduction While autoimmune pancreatitis was initially recognized in the 1960s, it was only in the 1990s that this inflammatory disease received substantial attention. The earliest comprehensive histologic description of this disease was from a group in Japan.1 However, it was the association of IgG4 with autoimmune pancreatitis, then referred to as sclerosing pancreatitis, that spurred new developments in this field.2 With increasing experience, the majority of patients with autoimmune pancreatitis are treated conservatively. Notably, pancreatic resections were the norm in the 20th century. In a fairly dramatic shift over the last decade, the diagnosis is frequently based on the presence of classic presenting symptoms, characteristic radiologic appearance, elevated serum IgG4 levels, as well as a swift response to
Rajib Gupta M.D. Post-doctoral Clinical Fellow, Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, MD, USA. Conflicts of interest: none declared. Vikram Deshpande M.D. Department of Pathology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA. Conflicts of interest: none declared.
DIAGNOSTIC HISTOPATHOLOGY --:-
1
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001
PANCREATIC PATHOLOGY
Histopathologic features to differentiate between type 1 and type 2 autoimmune pancreatitis Histologic features
Type 1 AIP
I. Around ducts: a. Lymphoplasmacytic collar Usually mild around the pancreatic ducts b. Ductal infiltration by neutrophils (granulocytic epithelial lesions), which may include duct-ulcer and/ or duct-abscess II. In the stroma: a. Storiform or cartwheel fibrosis b. Eosinophils c. Lymphoid follicles d. Obliterative Phlebitis and/ or arteritis III. Neutrophils within pancreatic acini IV. Immunohistochemistry: a. IgG4 in the stroma
b. IgG4:IgG ratio
Type 2 AIP
Rare to absent
Usually moderate to marked Present
Present
Absent
Common Sometimes seen Present
Rare Rare to absent
Rare
Common
Elevated, biopsy >10 per HPF, resection >50 per HPF >40%
Rare
Figure 2 Type 1 autoimmune pancreatitis involving the intrapancreatic portion of the bile duct. The biliary epithelium is eroded, likely a consequence of an indwelling stent. The full thickness involvement of the bile duct is a characteristic feature of this disease.
Absent
<40%
Table 1
Figure 3 Type 1 autoimmune pancreatitis. Note the periductal accentuation. Nevertheless, the ductal epithelium appears well preserved.
Figure 1 Type 1 autoimmune pancreatitis with prominent storiformtype fibrosis.
of the lumen) is viewed synonymously with total obliteration of the vein (Figures 4 and 5). It is important to emphasize that the pattern of obliteration differs significantly from organized thrombosis of a pancreatic vein, a finding seen in other forms of pancreatitis and malignancy. Unlike thrombosed veins, the obliterated veins in autoimmune pancreatitis show a
DIAGNOSTIC HISTOPATHOLOGY --:-
Figure 4 Type 1 autoimmune pancreatitis with obliterative phlebitis. A portion of this vein is completely obliterated (arrow), and the other segment is partially obliterated (arrow head).
2
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001
PANCREATIC PATHOLOGY
positive plasma cells on a biopsy, and greater than 50 IgG4 positive plasma cells on a high-power field on a resected pancreas delivers the ancillary evidence necessary to seal a diagnosis of autoimmune pancreatitis (Figure 7).3,7,8,15 The IgG4 positive plasma cells are distributed diffusely throughout the pancreas, although foci relatively devoid of IgG4 cells are invariably found. It should be emphasized that isolated elevations of IgG4 positive plasma cells are seen in a variety of inflammatory diseases of the pancreas, and of greater concern, an invasive ductal adenocarcinoma. An elevated IgG4 to IgG ratio provides an additional level of support to the diagnosis.7 However, immunoglobulin stains are generally associated with significant non-specific background signal, making computing a ratio a challenge. An in situ hybridization approach is often able to overcome these problems, and commercial probes for IgG4 and IgG on automated platforms are now increasingly available.16
Figure 5 Type 1 autoimmune pancreatitis with total obliteration of a vein (arrow). Such totally obliterated veins can be easily overlooked. Within the pancreas, veins run in parallel with arterial channels, and a well-circumscribed aggregate of inflammatory cells adjacent to an arterial channel should be carefully evaluated for the possibility of a obliterated venous channel.
Analysis of biopsy samples and clinical algorithm for diagnosis of type 1 autoimmune pancreatitis The differential diagnosis in patients with suspected autoimmune pancreatitis includes pancreatic adenocarcinoma, cholangiocarcinoma, idiopathic pancreatitis and primary sclerosing cholangitis. The primary objective in the vast majority of cases is to rule out malignancy. More than half of all patients with type 1 autoimmune pancreatitis can be diagnosed on the basis of cross-sectional imaging, serology and the involvement of other organs, particularly those characteristically affected by IgG4 related disease, and a biopsy is considered redundant in this scenario. Pancreatic biopsies often represent a major challenge to the pathologist, but are of vital importance, given the need to spare these individuals the morbidity and mortality associated with a pancreatic resection (Figure 8). The need to identify collateral evidence in the form of imaging, clinical and serological findings cannot be over emphasized, particularly in situations in which the biopsy either underrepresents the lesion or appears nonrepresentative. There are several challenges associated with the interpretation of a needle biopsy: 1) While storiform-type fibrosis is often observed, obliterative phlebitis is seldom seen with a needle biopsy.
fibroinflammatory infiltrate that is histologically similar to that seen in the background pancreas, and the inflammatory infiltrate occupies the lumen of the vein and extends through the wall. A similar appearance may be seen in arterial channels, the socalled obliterative arteritis (Figure 6). Remarkably, in spite of the extensive obliteration of arteries and veins, necrosis is seldom seen. Necrotizing vasculitis is notably absent in this disease. Involvement of the intrapancreatic portion of the bile duct is fairly common, and the inflammatory infiltrate and storiform fibrosis involve both the inner as well as the outer half of the bile duct (Figure 2). Unfortunately, these characteristic histologic features are seldom captured on an endoscopic biopsy, and thus bile duct biopsies are seldom diagnostic for IgG4 related disease. Immunohistochemical features The disease is characterized by elevated numbers of IgG4 positive plasma cells. As a general rule, greater than 10 IgG4
Figure 6 Type 1 autoimmune pancreatitis with partially obliterated arterial channel-obliterative arteritis.
DIAGNOSTIC HISTOPATHOLOGY --:-
Figure 7 Type 1 autoimmune pancreatitis diffusely infiltrated by IgG4 positive plasma cells (immunohistochemical stain for IgG4).
3
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001
PANCREATIC PATHOLOGY
elevated serum IgG4, 3) other organ involvement, and 4) a favorable response to a trial of steroids. On contrast enhanced CT, most cases of autoimmune pancreatitis show diffuse pancreatic swelling, often with a hypoattenuating capsule like rim.4,5 The finding of a long stricture on endoscopic retrograde pancreatography involving more than one-third of the duct and concomitant lack of upstream dilatation from the stricture, as well as multifocal strictures, are all highly suggestive of autoimmune pancreatitis. In the late phase of the disease, the pancreas may be atrophic and associated with stone/calcifications. It should be noted that like tissue IgG4, elevated serum IgG4 is a relatively non-specific finding. In a hospital-based series, the positive predictive value of an elevated serum IgG4 for IgG4 related disease was 34%.18 However, in the setting of a compelling clinical picture for autoimmune pancreatitis e an elderly male with obstructive jaundice e the specificity of serum IgG4 is significantly higher. In an attempt to improve the specificity of the assay, current algorithms suggest a cutoff of greater than two times the upper limit of normal. Mild elevations of serum IgG4 (1 e2 times the upper limit of normal) have been observed in 15% of pancreatic carcinomas. Recent data have also questioned the purported high sensitivity of the assay: approximately 20e40% of cases of IgG4 related disease/autoimmune pancreatitis show normal or only slightly elevated levels of serum IgG4. The presence of a tumefactive lesion in other organs such as sclerosing cholangitis, renal disease, or orbital or submandibular lesions, is powerful evidence of IgG4 related disease. The involvement may be synchronous, or precede pancreatic involvement, and a review of prior biopsy material from the archives could both deliver a definitive diagnosis of autoimmune pancreatitis and avert an invasive pancreatic procedure. Reliance on any one single parameter is unwise, and a careful and deliberate analysis by a multidisciplinary group of physicians remains key to avoid underdiagnosis as well as overdiagnosis of this disease.
Figure 8 A needle core biopsy from an individual with type 1 autoimmune pancreatitis. Significant fibrosis is present as is a sparse lymphocytic infiltrate. The biopsy lack storiform-type fibrosis and obliterative phlebitis. The IgG4 stain (not shown here) showed 5 IgG4 positive plasma cells per high-power field. Nevertheless, in conjunction with radiologic appearance, these findings are compatible with a diagnosis of autoimmune pancreatitis type 1.
2) Approximately 10% of pancreatic ductal adenocarcinomas are associated with peritumoral fibrosis and increased numbers of IgG4 positive plasma cells. It should be emphasized that the incidence of pancreatic carcinoma is several fold higher than autoimmune pancreatitis. Conversely, IgG4 counts in autoimmune pancreatitis may be lower than 10 per high-power field, which in most cases is related to the relative patchy nature of the IgG4 infiltrate. 3) Endoscopic ultrasound guided core biopsies are not widely available, and some regions of the pancreas are inaccessible to a core biopsy. Micro-tissue samples from a fine needle aspiration biopsy may assist in the diagnosis, but are seldom diagnostic in and of themselves.17 The primary role of a fine needle aspiration biopsy is to rule out a pancreatic ductal adenocarcinoma. Fortunately, the biopsy constitutes only one part of a larger diagnostic algorithm. These diagnostic algorithms, now codified by an international group of experts, are widely used in clinical practice (Table 2).3 The cardinal clinical features of autoimmune pancreatitis are: 1) a characteristic radiologic appearance, 2)
Novel diagnostics in IgG4 related disease A number of novel tools are emerging that may assist in the diagnosis of IgG4 related disease and autoimmune pancreatitis. IgG4 related disease is characterized by the presence of dominant clones of IgG4 positive cells, as identified by next generation sequencing, and their presence may assist in the distinction of
Clinical, radiologic and serologic features that assist in the histologic diagnoses of autoimmune pancreatitis. The presence of two or more of these features provides strong support to the histologic diagnoses of autoimmune pancreatitis
Radiology
Serum IgG4 Other organ involvement Response to steroids
Type 1 autoimmune pancreatitis
Type 2 autoimmune pancreatitis
Irregular narrowing of the main pancreatic duct and pancreatic swelling with a capsule like rim Elevated Liver, kidney, lymph nodes, submandibular salivary gland, lung Swift response to steroids
Irregular narrowing of the main pancreatic duct and pancreatic swelling with a capsule like rim Not elevated Inflammatory bowel disease, typically ulcerative colitis Swift response to steroids
Table 2
DIAGNOSTIC HISTOPATHOLOGY --:-
4
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001
PANCREATIC PATHOLOGY
autoimmune pancreatitis and IgG4 related sclerosing cholangitis from primary sclerosing cholangitis and pancreatic malignancy.19 Recent data from the authors’ institution reveal that the majority of cases with IgG4 related disease show elevated numbers of plasmablasts as well as an expansion of CD4 positive-effector T cells.20,21 Plasmablast numbers may also prove useful in the longitudinal management of IgG4 related disease. Although these assays have drawn considerable interest, their clinical value remains to be established. IgG4 negative IgG4 related disease? Autoimmune pancreatitis is widely considered to be an immunologically mediated disease, although the precise abnormalities are currently unknown. However, it is widely agreed that IgG4 may not the “driver” and instead may simply represent an epiphenomenon. IgG4 is considered a noninflammatory antibody because of its inability to fix complements and poor binding with Fc receptors. Thus, it should come as no surprise that reports of autoimmune pancreatitis and IgG4 related disease lacking IgG4 positive plasma cells have emerged in the last few years.22 The diagnosis in these instances relies on histologic features, the presence of multiorgan involvement, as well as a robust response to immunosuppressive therapy. Nevertheless, these cases are a relatively uncommon, and such diagnoses are best arrived at after a thorough evaluation by a multidisciplinary team.
Figure 9 Type 2 autoimmune pancreatitis. Note the dense periductal lymphocytic infiltrate. Marked pancreatic atrophy is present. Although fibrosis is evident, storiform-type fibrosis is absent. Granulocytic epithelial lesions are present, although they are difficult to discern at this low magnification.
Type 2 autoimmune pancreatitis Type 2 autoimmune pancreatitis is typically identified in younger individuals (mean age 30e40 years), an uncommon age for pancreatic ductal adenocarcinoma.4 Unlike type 1 autoimmune pancreatitis, these patients frequently present with acute pancreatitis, although a pancreatic mass and/or obstructive jaundice is also seen in a substantial number of patients.4 Little is known about the pathogenic mechanisms of type 2 autoimmune pancreatitis, although given the histologic appearance, an immune-mediated disease is considered likely.
Figure 10 Type 2 autoimmune pancreatitis with granulocytic epithelial lesions.
Histopathology The disease is characterized by the presence of a robust periductal inflammatory infiltrate and granulocytic epithelial lesions, the latter best illustrated by the presence of abscesses within ductal epithelium (Table 1).9,13e15 The periductal infiltrate and the accompanying neutrophils involve both large-caliber pancreatic ducts as well as more peripheral ducts (Figures 9 and 10). The periductal infiltrate is composed predominantly of lymphocytes and plasma cells. Unlike type 1 autoimmune pancreatitis, the ductal epithelium shows considerable damage, and in the most extreme examples show total denudation of epithelium. Reactive changes of ductal epithelial cells are often observed, although these findings are seldom concerning for a neoplastic process. The neutrophilic infiltrate also tends to extend into the lobules, and neutrophils within acinar structures are frequently observed (Figure 11).13e15,23 Occasional granulomas are identified, although the presence of widespread periductal granulomas should raise the possibility of an alternative form of granulomatous pancreatitis.
DIAGNOSTIC HISTOPATHOLOGY --:-
Figure 11 Type 2 autoimmune pancreatitis with a sparse neutrophilic infiltrate involving pancreatic acini.
5
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001
PANCREATIC PATHOLOGY
pancreatic neoplasm. The presence of inflammation localized to the region of the ‘groove’ between the bile duct and the duodenum should raise the possibility of groove pancreatitis.24 However, in the author’s experience, extensive granulocytic epithelial lesion formation accompanied by a brisk periductal lymphoplasmacytic infiltrate is an uncommon finding in groove pancreatitis. The stigmata of alcoholic pancreatitis, dilated ducts filled with inspissated material, are not seen in type 2 autoimmune pancreatitis. Intraductal and acinar neutrophils may be seen adjacent to pancreatic ductal adenocarcinoma, and this finding represents a potential pitfall, particularly when evaluating needle biopsy specimens. Follicular pancreatitis may also mimic both type 1 and type 2 autoimmune pancreatitis.25 This disease is characterized by the presence of periductal lymphoid aggregates, however, granulocytic epithelial lesions and increased numbers of IgG4 positive cells are not seen.
The adjacent pancreas invariably shows significant atrophy, likely a consequence of ductal obstruction. Both interlobular and intralobular fibrosis is noted, however, the character of fibrosis is unlike that seen in type 1 autoimmune pancreatitis, lacking storiform fibrosis, as well as the cellular inflamed stroma, both hallmarks of type 1 autoimmune pancreatitis. Foci of phlebitis may be observed, however, obliterative phlebitis is generally absent. Immunohistochemistry Type 2 autoimmune pancreatitis is not an IgG4 related disease, and elevated numbers of IgG4 positive plasma cells are distinctly uncommon.13e15,23 Nevertheless, particularly on a resected specimen, occasional clusters of plasma cells rich in IgG4þ cells may be identified. The IgG4 to IgG ratio is generally less than 40%. Some cases of autoimmune pancreatitis show features of both type 1 and type 2 disease. The most common scenario is one in which the histologic and immunohistochemical findings are compatible with type 1 autoimmune pancreatitis with one exception, occasional neutrophils and intraductal neutrophilic aggregates are present. In our experience, these cases are best classified as type 1 autoimmune pancreatitis. Nevertheless, classification of some cases may be legitimately deferred, and the term unclassified autoimmune pancreatitis may be justified in certain instances.
Analysis of biopsy samples and clinical algorithm for the diagnosis of type 2 autoimmune pancreatitis As with type 1 autoimmune pancreatitis, the results of a biopsy cannot be viewed in isolation, instead should be evaluated in the broader clinical and radiological context (Table 2).4,6,26,27 However, unlike type 1 disease, type 2 autoimmune pancreatitis is confined to the pancreas, shows normal serum IgG4 levels, and thus neither of these parameters help in distinguishing this disease from pancreatic carcinoma. The imaging findings are similar to type 1 disease. The age of the patient is of considerable help, the majority of patients are younger than 50 years, a relatively uncommon demographic for pancreatic carcinoma. A strong association exists between type 1 autoimmune pancreatitis and inflammatory bowel disease: one-third of cases in the authors’ series were associated with inflammatory bowel disease. In summary, the two most robust pieces of corroborative evidence for the histologic diagnosis of type 2 autoimmune pancreatitis are: 1) radiologic findings, and 2) a history of inflammatory bowel disease.
Differential diagnosis Histology represents the single best gold standard for the diagnosis of type 2 autoimmune pancreatitis, and unlike type 1 autoimmune pancreatitis, there is no biomarker for this disease. In and of itself, granulocytic epithelial lesions are not diagnostic of type 2 autoimmune pancreatitis (Figure 12). Neutrophilic aggregates may be identified with diseases such as alcoholic and other forms of acute and chronic pancreatitis, groove pancreatitis, and most significantly, adjacent to a
Malignancy and autoimmune pancreatitis There is considerable uncertainty about the risk of malignancy in individuals with autoimmune pancreatitis. Bona fide cases of autoimmune pancreatitis progressing to malignancy have been documented.28,29 Low-grade PanIN is ubiquitous: in one study 82% of autoimmune pancreatitis cases showed grade 1 PanIN while 25% showed grade 2 PanIN.28 The advanced age of patients at the time of diagnosis, the relatively low incidence of autoimmune pancreatitis, as well as the lack of long-term followup in most series, precludes a thorough assessment of the risk of malignancy in autoimmune pancreatitis. A recent study identified a disproportionate increase in cancer in the first year after diagnosis of autoimmune pancreatitis, raising the intriguing possibility that autoimmune pancreatitis, in some cases, may represent a paraneoplastic phenomenon.29
Figure 12 Type 2 autoimmune pancreatitis. A needle core biopsy shows fibrosis (left) and pancreatic acinar tissue infiltrated by a neutrophilic infiltrate. An immunohistochemical stain for IgG4 is negative (not shown). Although in and of itself this appearance is not diagnostic of type 2 autoimmune pancreatitis, in conjunction with the radiologic appearance and a history of ulcerative colitis, the findings are compatible with a diagnosis of type 2 autoimmune pancreatitis.
DIAGNOSTIC HISTOPATHOLOGY --:-
Ampullary biopsy for the diagnosis of autoimmune pancreatitis Ampullary biopsies are often used to support a diagnosis of autoimmune pancreatitis: patients with autoimmune
6
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001
PANCREATIC PATHOLOGY
pancreatitis typically show 10 IgG4 positive plasma cells per high-power field.30,31 It should be emphasized that the histologic hallmarks of autoimmune pancreatitis are seldom seen in an ampullary biopsy, and hence this assay relies solely on identifying elevated numbers of IgG4 positive plasma cells, a less than optimal approach since elevated numbers of IgG4 positive plasma cells are not specific to autoimmune pancreatitis. Ampullary biopsies from individuals with type 2 autoimmune pancreatitis are generally negative for IgG4 positive plasma cells. In summary, although in some circumstances ampullary biopsies may be informative, it is best not to rely too heavily on these results.
4 Hart PA, Levy MJ, Smyrk TC, et al. Clinical profiles and outcomes in idiopathic duct-centric chronic pancreatitis (type 2 autoimmune pancreatitis): the Mayo Clinic experience. Gut 2015. gutjnl-2015309275, [Epub ahead of print]. 5 Kamisawa T, Chari ST, Lerch MM, et al. Recent advances in autoimmune pancreatitis: type 1 and type 2. Gut 2013; 62: 1373e80. 6 Okazaki K, Uchida K. Autoimmune pancreatitis: the past, present, and future. Pancreas 2015; 44: 1006e16. 7 Deshpande V, Zen Y, Chan JK, et al. Consensus statement on the pathology of IgG4-related disease. Mod Pathol 2012; 25: 1181e92. 8 Zhang L, Chari S, Smyrk TC, et al. Autoimmune pancreatitis (AIP) type 1 and type 2: an international consensus study on histopathologic diagnostic criteria. Pancreas 2011; 40: 1172e9. 9 Deshpande V, Chicano S, Finkelberg D, et al. Autoimmune pancreatitis: a systemic immune complex mediated disease. Am J Surg Pathol 2006; 30: 1537e45. 10 Deshpande V, Mino-Kenudson M, Brugge W, et al. Autoimmune pancreatitis: more than just a pancreatic disease? A contemporary review of its pathology. Arch Pathol Lab Med 2005; 129: 1148e54. 11 Mahajan VS, Mattoo H, Deshpande V, et al. IgG4-related disease. Annu Rev Pathol 2014; 9: 315e47. 12 Stone JH, Zen Y, Deshpande V. IgG4-related disease. N Engl J Med 2012; 366: 539e51. 13 Notohara K, Burgart LJ, Yadav D, et al. Idiopathic chronic pancreatitis with periductal lymphoplasmacytic infiltration: clinicopathologic features of 35 cases. Am J Surg Pathol 2003; 27: 1119e27. 14 Shinagare S, Shinagare AB, Deshpande V. Autoimmune pancreatitis: a guide for the histopathologist. Semin Diagn Pathol 2012; 29: 197e204. 15 Deshpande V, Gupta R, Sainani N, et al. Subclassification of autoimmune pancreatitis: a histologic classification with clinical significance. Am J Surg Pathol 2011; 35: 26e35. 16 Tse J, Riggi N, Ting D, et al. Diagnosis of IgG4-related disease: no longer an IS (H)-Sue? Lab Invest 2013. 184A-184A. 17 Deshpande V, Mino-Kenudson M, Brugge WR, et al. Endoscopic ultrasound guided fine needle aspiration biopsy of autoimmune pancreatitis: diagnostic criteria and pitfalls. Am J Surg Pathol 2005; 29: 1464e71. 18 Carruthers MN, Khosroshahi A, Augustin T, et al. The diagnostic utility of serum IgG4 concentrations in IgG4-related disease. Ann Rheum Dis 2015; 74: 14e8. 19 Doorenspleet ME, Hubers LM, Culver EL, et al. IgG4þ B-cell receptor clones distinguish IgG4 related disease from primary sclerosing cholangitis and biliary/pancreatic malignancies. Hepatology 2016 [Epub ahead of print]. 20 Wallace ZS, Mattoo H, Carruthers M, et al. Plasmablasts as a biomarker for IgG4-related disease, independent of serum IgG4 concentrations. Ann Rheum Dis 2015; 74: 190e5. 21 Mattoo H, Mahajan VS, Maehara T, et al. Clonal expansion of CD4 cytotoxic T lymphocytes in patients with IgG-related disease. J Allergy Clin Immunol 2016 [Epub ahead of print]. 22 Hart PA, Smyrk TC, Chari ST. Lymphoplasmacytic sclerosing pancreatitis without IgG4 tissue infiltration or serum IgG4 elevation: IgG4-related disease without IgG4. Mod Pathol 2015; 28: 238e47.
Treatment Steroids are the standard of care for the treatment of autoimmune pancreatitis. High-dose oral prednisone at a dose of 0.6e1 mg per kg per day is administered for 3e4 weeks.4,6,11 A relief of pancreatic swelling and symptoms can be observed as early as two weeks. Approximately half of all patients with type 1 autoimmune pancreatitis relapse, while those with type 2 disease seldom relapse, and recurrence with the latter disease is confined to the pancreas. There is no consensus about the duration of steroid treatment, and the benefit of maintenance steroids continues to be debated. In cases with relapse, three treatment regimens have been used: 1) high-dose steroids followed by maintenance therapy, 2) high-dose steroids without maintenance therapy, and 3) rituximab induction with or without maintenance rituximab. Azathioprine, 6-mercaptopurine, mycofenolate mofetil (MMF) have been used as steroid sparing agents for maintenance therapy. Rituximab has been shown to be particularly effective in cases that are resistant to steroid therapy and as a steroid-sparing agent.32e34
Conclusions The diagnostic approach of autoimmune pancreatitis relies on the coexistence of supportive clinical, laboratory, radiologic and histologic findings, none of which in and of themselves are pathognomonic. Nevertheless, histology represents the single best gold standard for a diagnosis of IgG4 related disease. The ultimate goal of the team of physicians is to avoid the morbidity and mortality of a pancreatic resection in individuals with autoimmune pancreatitis. Prompt additional investigations to exclude malignancy should be undertaken in individuals that lack a swift response to immunosuppressive therapy. A REFERENCES 1 Kawaguchi K, Koike M, Tsuruta K, et al. Lymphoplasmacytic sclerosing pancreatitis with cholangitis: a variant of primary sclerosing cholangitis extensively involving pancreas. Hum Pathol 1991; 22: 387e95. 2 Hamano H, Kawa S, Horiuchi A, et al. High serum IgG4 concentrations in patients with sclerosing pancreatitis. N Engl J Med 2001; 344: 732e8. 3 Shimosegawa T, Chari ST, Frulloni L, et al. International consensus diagnostic criteria for autoimmune pancreatitis: guidelines of the International Association of Pancreatology. Pancreas 2011; 40: 352e8.
DIAGNOSTIC HISTOPATHOLOGY --:-
7
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001
PANCREATIC PATHOLOGY
23 Zamboni G, Luttges J, Capelli P, et al. Histopathological features of diagnostic and clinical relevance in autoimmune pancreatitis: a study on 53 resection specimens and 9 biopsy specimens. Virchows Arch 2004; 445: 552e63. 24 Raman SP, Salaria SN, Hruban RH, et al. Groove pancreatitis: spectrum of imaging findings and radiology-pathology correlation. AJR Am J Roentgenol 2013; 201: W29e39. 25 Gupta RK, Xie BH, Patton KT, et al. Follicular pancreatitis: a distinct form of chronic pancreatitis-an additional mimic of pancreatic neoplasms. Hum Pathol 2016; 48: 154e62. 26 Ueki T, Kawamoto K, Otsuka Y, et al. Prevalence and clinicopathological features of autoimmune pancreatitis in Japanese patients with inflammatory bowel disease. Pancreas 2015; 44: 434e40. 27 Notohara K, Nishimori I, Mizuno N, et al. Clinicopathological features of type 2 autoimmune pancreatitis in japan: results of a multicenter survey. Pancreas 2015; 44: 1072e7. 28 Gupta R, Khosroshahi A, Shinagare S, et al. Does autoimmune pancreatitis increase the risk of pancreatic carcinoma?: a retrospective analysis of pancreatic resections. Pancreas 2013; 42: 506e10.
DIAGNOSTIC HISTOPATHOLOGY --:-
29 Shiokawa M, Kodama Y, Yoshimura K, et al. Risk of cancer in patients with autoimmune pancreatitis. Am J Gastroenterol 2013; 108: 610e7. 30 Kawakami H, Zen Y, Kuwatani M, et al. IgG4-related sclerosing cholangitis and autoimmune pancreatitis: histological assessment of biopsies from Vater’s ampulla and the bile duct. J Gastroenterol Hepatol 2010; 25: 1648e55. 31 Sepehr A, Mino-Kenudson M, Ogawa F, et al. IgG4þ to IgGþ plasma cells ratio of ampulla can help differentiate autoimmune pancreatitis from other “mass forming” pancreatic lesions. Am J Surg Pathol 2008; 32: 1770e9. 32 Khosroshahi A, Carruthers MN, Deshpande V, et al. Rituximab for the treatment of IgG4-related disease: lessons from 10 consecutive patients. Medicine (Baltimore) 2012; 91: 57e66. 33 Khosroshahi A, Bloch DB, Deshpande V, et al. Rituximab therapy leads to rapid decline of serum IgG4 levels and prompt clinical improvement in IgG4-related systemic disease. Arthritis Rheum 2010; 62: 1755e62. 34 Hart PA, Topazian MD, Witzig TE, et al. Treatment of relapsing autoimmune pancreatitis with immunomodulators and rituximab: the Mayo Clinic experience. Gut 2013; 62: 1607e15.
8
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001
immunosuppressive therapy.3e5 Nevertheless, in some cases the disease shows overlapping features with pancreatic cancer, and it is in these cases that histopathology plays a critical role. Additionally, the lack of a specific biomarker for type 2 autoimmune pancreatitis (see below) has meant that these patients are frequently subjected to a biopsy, and occasionally, a pancreatic resection.
Current challenges in the diagnosis of autoimmune pancreatitis Rajib Gupta
Variants of autoimmune pancreatitis
Vikram Deshpande
Two distinct variants of autoimmune pancreatitis are recognized, labeled type 1 and type 2.5e10 The two variants are clinically, serologically, as well as histologically distinctive, although both forms of the disease respond to immunosuppressive therapy. In fact, based on these differences, it has been suggested that the two “variants” represent distinct diseases, a position that we endorse. The concept of type 1 autoimmune pancreatitis was originally proposed in Japan, while type 2 autoimmune pancreatitis was championed in Europe, the latter primarily on the basis of histopathologic findings. Notably, type 2 autoimmune pancreatitis is uncommon in the Asian population.
Abstract Autoimmune pancreatitis (AIP), an inflammatory disease of the pancreas first described in Japan, is characterized by a tumefactive lesion in the pancreas, thus clinically resembling pancreatic cancer. Currently, it is classified into two variants e type 1 AIP and type 2 AIP, each with distinct clinico-pathologic and epidemiologic features. Type 1 AIP is a part of IgG4 related disease, a generalized multi-systemic disease, and recapitulates its clinical and pathologic features. Type 2 AIP is an isolated pancreas-centric disease, and is not associated with elevated tissue or serum IgG4. Histologically, type 1 AIP is characterized by storiform-type fibrosis and obliterative phlebitis, while type 2 disease shows granulocytic epithelial lesions. Even when AIP is suspected, a diagnosis should seldom be made in isolation, but should be rendered only after collating clinical and radiologic data. Steroids remain the mainstay of treatment for both variants of AIP, although other immunosuppressive drugs like rituximab have proven successful. In this review, we summarize the current knowledge of this entity and discuss diagnostic challenges, with particular emphasis on the interpretation of needle biopsies. Recognition of AIP as a mimic of pancreatic cancer is imperative to reduce unnecessary morbidity and mortality associated with pancreatic surgery.
Type 1 autoimmune pancreatitis This is the pancreatic manifestation of IgG4 related disease. IgG4 related disease is a multi-systemic disease characterized by the presence of multifocal fibroinflammatory mass lesion(s) that are responsive to immunosuppressive therapy.11,12 Type 1 autoimmune pancreatitis bears all the clinical, serologic, and histologic hallmarks of IgG4 related disease. This is a disease of elderly men and the majority of these patients (60e75%) present with painless obstructive jaundice. A small proportion present with acute pancreatitis. Rarely the patient is entirely asymptomatic, the disease detected on cross-sectional imaging.
Keywords autoimmune pancreatitis; IgG4; IgG4 related disease
Histopathology Virtually all pancreatic resections from individuals with type 1 autoimmune pancreatitis reveal a dense lymphoplasmacytic infiltrate, storiform-type fibrosis and obliterative phlebitis (Table 1).8,9,13,14 In the authors’ experience, only patients pre-treated with immunosuppressive therapy lack a brisk inflammatory infiltrate. The infiltrate is generally diffusely distributed throughout the pancreas, although accentuation in the periductal region is often appreciable (Figures 1e3). Interestingly, although significant inflammation surrounds pancreatic ducts, the ductal epithelium is well preserved. An increased number of eosinophils are invariably found, and in rare instances eosinophils may dominate the histologic picture, and in numbers sufficient to raise the possibility of an allergic or an unrelated connective tissue disease. The disease is characterized by dense fibrosis, generally laid down in a storiform pattern (Figure 1). This is a pattern familiar to pathologists, most notably seen in dermatofibrosarcoma protuberans. The pattern is likely fashioned by the deposition of fibroblasts or myofibroblasts, the latter representing a consistent finding in autoimmune pancreatitis. Hyalinizing or keloidal type fibrosis is uncommon in this variant of autoimmune pancreatitis. Perhaps the most unique feature of autoimmune pancreatitis is obliterative phlebitis. Partial obliteration (>obliteration of 50%
Introduction While autoimmune pancreatitis was initially recognized in the 1960s, it was only in the 1990s that this inflammatory disease received substantial attention. The earliest comprehensive histologic description of this disease was from a group in Japan.1 However, it was the association of IgG4 with autoimmune pancreatitis, then referred to as sclerosing pancreatitis, that spurred new developments in this field.2 With increasing experience, the majority of patients with autoimmune pancreatitis are treated conservatively. Notably, pancreatic resections were the norm in the 20th century. In a fairly dramatic shift over the last decade, the diagnosis is frequently based on the presence of classic presenting symptoms, characteristic radiologic appearance, elevated serum IgG4 levels, as well as a swift response to
Rajib Gupta M.D. Post-doctoral Clinical Fellow, Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, MD, USA. Conflicts of interest: none declared. Vikram Deshpande M.D. Department of Pathology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA. Conflicts of interest: none declared.
DIAGNOSTIC HISTOPATHOLOGY --:-
1
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001
PANCREATIC PATHOLOGY
Histopathologic features to differentiate between type 1 and type 2 autoimmune pancreatitis Histologic features
Type 1 AIP
I. Around ducts: a. Lymphoplasmacytic collar Usually mild around the pancreatic ducts b. Ductal infiltration by neutrophils (granulocytic epithelial lesions), which may include duct-ulcer and/ or duct-abscess II. In the stroma: a. Storiform or cartwheel fibrosis b. Eosinophils c. Lymphoid follicles d. Obliterative Phlebitis and/ or arteritis III. Neutrophils within pancreatic acini IV. Immunohistochemistry: a. IgG4 in the stroma
b. IgG4:IgG ratio
Type 2 AIP
Rare to absent
Usually moderate to marked Present
Present
Absent
Common Sometimes seen Present
Rare Rare to absent
Rare
Common
Elevated, biopsy >10 per HPF, resection >50 per HPF >40%
Rare
Figure 2 Type 1 autoimmune pancreatitis involving the intrapancreatic portion of the bile duct. The biliary epithelium is eroded, likely a consequence of an indwelling stent. The full thickness involvement of the bile duct is a characteristic feature of this disease.
Absent
<40%
Table 1
Figure 3 Type 1 autoimmune pancreatitis. Note the periductal accentuation. Nevertheless, the ductal epithelium appears well preserved.
Figure 1 Type 1 autoimmune pancreatitis with prominent storiformtype fibrosis.
of the lumen) is viewed synonymously with total obliteration of the vein (Figures 4 and 5). It is important to emphasize that the pattern of obliteration differs significantly from organized thrombosis of a pancreatic vein, a finding seen in other forms of pancreatitis and malignancy. Unlike thrombosed veins, the obliterated veins in autoimmune pancreatitis show a
DIAGNOSTIC HISTOPATHOLOGY --:-
Figure 4 Type 1 autoimmune pancreatitis with obliterative phlebitis. A portion of this vein is completely obliterated (arrow), and the other segment is partially obliterated (arrow head).
2
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001
PANCREATIC PATHOLOGY
positive plasma cells on a biopsy, and greater than 50 IgG4 positive plasma cells on a high-power field on a resected pancreas delivers the ancillary evidence necessary to seal a diagnosis of autoimmune pancreatitis (Figure 7).3,7,8,15 The IgG4 positive plasma cells are distributed diffusely throughout the pancreas, although foci relatively devoid of IgG4 cells are invariably found. It should be emphasized that isolated elevations of IgG4 positive plasma cells are seen in a variety of inflammatory diseases of the pancreas, and of greater concern, an invasive ductal adenocarcinoma. An elevated IgG4 to IgG ratio provides an additional level of support to the diagnosis.7 However, immunoglobulin stains are generally associated with significant non-specific background signal, making computing a ratio a challenge. An in situ hybridization approach is often able to overcome these problems, and commercial probes for IgG4 and IgG on automated platforms are now increasingly available.16
Figure 5 Type 1 autoimmune pancreatitis with total obliteration of a vein (arrow). Such totally obliterated veins can be easily overlooked. Within the pancreas, veins run in parallel with arterial channels, and a well-circumscribed aggregate of inflammatory cells adjacent to an arterial channel should be carefully evaluated for the possibility of a obliterated venous channel.
Analysis of biopsy samples and clinical algorithm for diagnosis of type 1 autoimmune pancreatitis The differential diagnosis in patients with suspected autoimmune pancreatitis includes pancreatic adenocarcinoma, cholangiocarcinoma, idiopathic pancreatitis and primary sclerosing cholangitis. The primary objective in the vast majority of cases is to rule out malignancy. More than half of all patients with type 1 autoimmune pancreatitis can be diagnosed on the basis of cross-sectional imaging, serology and the involvement of other organs, particularly those characteristically affected by IgG4 related disease, and a biopsy is considered redundant in this scenario. Pancreatic biopsies often represent a major challenge to the pathologist, but are of vital importance, given the need to spare these individuals the morbidity and mortality associated with a pancreatic resection (Figure 8). The need to identify collateral evidence in the form of imaging, clinical and serological findings cannot be over emphasized, particularly in situations in which the biopsy either underrepresents the lesion or appears nonrepresentative. There are several challenges associated with the interpretation of a needle biopsy: 1) While storiform-type fibrosis is often observed, obliterative phlebitis is seldom seen with a needle biopsy.
fibroinflammatory infiltrate that is histologically similar to that seen in the background pancreas, and the inflammatory infiltrate occupies the lumen of the vein and extends through the wall. A similar appearance may be seen in arterial channels, the socalled obliterative arteritis (Figure 6). Remarkably, in spite of the extensive obliteration of arteries and veins, necrosis is seldom seen. Necrotizing vasculitis is notably absent in this disease. Involvement of the intrapancreatic portion of the bile duct is fairly common, and the inflammatory infiltrate and storiform fibrosis involve both the inner as well as the outer half of the bile duct (Figure 2). Unfortunately, these characteristic histologic features are seldom captured on an endoscopic biopsy, and thus bile duct biopsies are seldom diagnostic for IgG4 related disease. Immunohistochemical features The disease is characterized by elevated numbers of IgG4 positive plasma cells. As a general rule, greater than 10 IgG4
Figure 6 Type 1 autoimmune pancreatitis with partially obliterated arterial channel-obliterative arteritis.
DIAGNOSTIC HISTOPATHOLOGY --:-
Figure 7 Type 1 autoimmune pancreatitis diffusely infiltrated by IgG4 positive plasma cells (immunohistochemical stain for IgG4).
3
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001
PANCREATIC PATHOLOGY
elevated serum IgG4, 3) other organ involvement, and 4) a favorable response to a trial of steroids. On contrast enhanced CT, most cases of autoimmune pancreatitis show diffuse pancreatic swelling, often with a hypoattenuating capsule like rim.4,5 The finding of a long stricture on endoscopic retrograde pancreatography involving more than one-third of the duct and concomitant lack of upstream dilatation from the stricture, as well as multifocal strictures, are all highly suggestive of autoimmune pancreatitis. In the late phase of the disease, the pancreas may be atrophic and associated with stone/calcifications. It should be noted that like tissue IgG4, elevated serum IgG4 is a relatively non-specific finding. In a hospital-based series, the positive predictive value of an elevated serum IgG4 for IgG4 related disease was 34%.18 However, in the setting of a compelling clinical picture for autoimmune pancreatitis e an elderly male with obstructive jaundice e the specificity of serum IgG4 is significantly higher. In an attempt to improve the specificity of the assay, current algorithms suggest a cutoff of greater than two times the upper limit of normal. Mild elevations of serum IgG4 (1 e2 times the upper limit of normal) have been observed in 15% of pancreatic carcinomas. Recent data have also questioned the purported high sensitivity of the assay: approximately 20e40% of cases of IgG4 related disease/autoimmune pancreatitis show normal or only slightly elevated levels of serum IgG4. The presence of a tumefactive lesion in other organs such as sclerosing cholangitis, renal disease, or orbital or submandibular lesions, is powerful evidence of IgG4 related disease. The involvement may be synchronous, or precede pancreatic involvement, and a review of prior biopsy material from the archives could both deliver a definitive diagnosis of autoimmune pancreatitis and avert an invasive pancreatic procedure. Reliance on any one single parameter is unwise, and a careful and deliberate analysis by a multidisciplinary group of physicians remains key to avoid underdiagnosis as well as overdiagnosis of this disease.
Figure 8 A needle core biopsy from an individual with type 1 autoimmune pancreatitis. Significant fibrosis is present as is a sparse lymphocytic infiltrate. The biopsy lack storiform-type fibrosis and obliterative phlebitis. The IgG4 stain (not shown here) showed 5 IgG4 positive plasma cells per high-power field. Nevertheless, in conjunction with radiologic appearance, these findings are compatible with a diagnosis of autoimmune pancreatitis type 1.
2) Approximately 10% of pancreatic ductal adenocarcinomas are associated with peritumoral fibrosis and increased numbers of IgG4 positive plasma cells. It should be emphasized that the incidence of pancreatic carcinoma is several fold higher than autoimmune pancreatitis. Conversely, IgG4 counts in autoimmune pancreatitis may be lower than 10 per high-power field, which in most cases is related to the relative patchy nature of the IgG4 infiltrate. 3) Endoscopic ultrasound guided core biopsies are not widely available, and some regions of the pancreas are inaccessible to a core biopsy. Micro-tissue samples from a fine needle aspiration biopsy may assist in the diagnosis, but are seldom diagnostic in and of themselves.17 The primary role of a fine needle aspiration biopsy is to rule out a pancreatic ductal adenocarcinoma. Fortunately, the biopsy constitutes only one part of a larger diagnostic algorithm. These diagnostic algorithms, now codified by an international group of experts, are widely used in clinical practice (Table 2).3 The cardinal clinical features of autoimmune pancreatitis are: 1) a characteristic radiologic appearance, 2)
Novel diagnostics in IgG4 related disease A number of novel tools are emerging that may assist in the diagnosis of IgG4 related disease and autoimmune pancreatitis. IgG4 related disease is characterized by the presence of dominant clones of IgG4 positive cells, as identified by next generation sequencing, and their presence may assist in the distinction of
Clinical, radiologic and serologic features that assist in the histologic diagnoses of autoimmune pancreatitis. The presence of two or more of these features provides strong support to the histologic diagnoses of autoimmune pancreatitis
Radiology
Serum IgG4 Other organ involvement Response to steroids
Type 1 autoimmune pancreatitis
Type 2 autoimmune pancreatitis
Irregular narrowing of the main pancreatic duct and pancreatic swelling with a capsule like rim Elevated Liver, kidney, lymph nodes, submandibular salivary gland, lung Swift response to steroids
Irregular narrowing of the main pancreatic duct and pancreatic swelling with a capsule like rim Not elevated Inflammatory bowel disease, typically ulcerative colitis Swift response to steroids
Table 2
DIAGNOSTIC HISTOPATHOLOGY --:-
4
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001
PANCREATIC PATHOLOGY
autoimmune pancreatitis and IgG4 related sclerosing cholangitis from primary sclerosing cholangitis and pancreatic malignancy.19 Recent data from the authors’ institution reveal that the majority of cases with IgG4 related disease show elevated numbers of plasmablasts as well as an expansion of CD4 positive-effector T cells.20,21 Plasmablast numbers may also prove useful in the longitudinal management of IgG4 related disease. Although these assays have drawn considerable interest, their clinical value remains to be established. IgG4 negative IgG4 related disease? Autoimmune pancreatitis is widely considered to be an immunologically mediated disease, although the precise abnormalities are currently unknown. However, it is widely agreed that IgG4 may not the “driver” and instead may simply represent an epiphenomenon. IgG4 is considered a noninflammatory antibody because of its inability to fix complements and poor binding with Fc receptors. Thus, it should come as no surprise that reports of autoimmune pancreatitis and IgG4 related disease lacking IgG4 positive plasma cells have emerged in the last few years.22 The diagnosis in these instances relies on histologic features, the presence of multiorgan involvement, as well as a robust response to immunosuppressive therapy. Nevertheless, these cases are a relatively uncommon, and such diagnoses are best arrived at after a thorough evaluation by a multidisciplinary team.
Figure 9 Type 2 autoimmune pancreatitis. Note the dense periductal lymphocytic infiltrate. Marked pancreatic atrophy is present. Although fibrosis is evident, storiform-type fibrosis is absent. Granulocytic epithelial lesions are present, although they are difficult to discern at this low magnification.
Type 2 autoimmune pancreatitis Type 2 autoimmune pancreatitis is typically identified in younger individuals (mean age 30e40 years), an uncommon age for pancreatic ductal adenocarcinoma.4 Unlike type 1 autoimmune pancreatitis, these patients frequently present with acute pancreatitis, although a pancreatic mass and/or obstructive jaundice is also seen in a substantial number of patients.4 Little is known about the pathogenic mechanisms of type 2 autoimmune pancreatitis, although given the histologic appearance, an immune-mediated disease is considered likely.
Figure 10 Type 2 autoimmune pancreatitis with granulocytic epithelial lesions.
Histopathology The disease is characterized by the presence of a robust periductal inflammatory infiltrate and granulocytic epithelial lesions, the latter best illustrated by the presence of abscesses within ductal epithelium (Table 1).9,13e15 The periductal infiltrate and the accompanying neutrophils involve both large-caliber pancreatic ducts as well as more peripheral ducts (Figures 9 and 10). The periductal infiltrate is composed predominantly of lymphocytes and plasma cells. Unlike type 1 autoimmune pancreatitis, the ductal epithelium shows considerable damage, and in the most extreme examples show total denudation of epithelium. Reactive changes of ductal epithelial cells are often observed, although these findings are seldom concerning for a neoplastic process. The neutrophilic infiltrate also tends to extend into the lobules, and neutrophils within acinar structures are frequently observed (Figure 11).13e15,23 Occasional granulomas are identified, although the presence of widespread periductal granulomas should raise the possibility of an alternative form of granulomatous pancreatitis.
DIAGNOSTIC HISTOPATHOLOGY --:-
Figure 11 Type 2 autoimmune pancreatitis with a sparse neutrophilic infiltrate involving pancreatic acini.
5
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001
PANCREATIC PATHOLOGY
pancreatic neoplasm. The presence of inflammation localized to the region of the ‘groove’ between the bile duct and the duodenum should raise the possibility of groove pancreatitis.24 However, in the author’s experience, extensive granulocytic epithelial lesion formation accompanied by a brisk periductal lymphoplasmacytic infiltrate is an uncommon finding in groove pancreatitis. The stigmata of alcoholic pancreatitis, dilated ducts filled with inspissated material, are not seen in type 2 autoimmune pancreatitis. Intraductal and acinar neutrophils may be seen adjacent to pancreatic ductal adenocarcinoma, and this finding represents a potential pitfall, particularly when evaluating needle biopsy specimens. Follicular pancreatitis may also mimic both type 1 and type 2 autoimmune pancreatitis.25 This disease is characterized by the presence of periductal lymphoid aggregates, however, granulocytic epithelial lesions and increased numbers of IgG4 positive cells are not seen.
The adjacent pancreas invariably shows significant atrophy, likely a consequence of ductal obstruction. Both interlobular and intralobular fibrosis is noted, however, the character of fibrosis is unlike that seen in type 1 autoimmune pancreatitis, lacking storiform fibrosis, as well as the cellular inflamed stroma, both hallmarks of type 1 autoimmune pancreatitis. Foci of phlebitis may be observed, however, obliterative phlebitis is generally absent. Immunohistochemistry Type 2 autoimmune pancreatitis is not an IgG4 related disease, and elevated numbers of IgG4 positive plasma cells are distinctly uncommon.13e15,23 Nevertheless, particularly on a resected specimen, occasional clusters of plasma cells rich in IgG4þ cells may be identified. The IgG4 to IgG ratio is generally less than 40%. Some cases of autoimmune pancreatitis show features of both type 1 and type 2 disease. The most common scenario is one in which the histologic and immunohistochemical findings are compatible with type 1 autoimmune pancreatitis with one exception, occasional neutrophils and intraductal neutrophilic aggregates are present. In our experience, these cases are best classified as type 1 autoimmune pancreatitis. Nevertheless, classification of some cases may be legitimately deferred, and the term unclassified autoimmune pancreatitis may be justified in certain instances.
Analysis of biopsy samples and clinical algorithm for the diagnosis of type 2 autoimmune pancreatitis As with type 1 autoimmune pancreatitis, the results of a biopsy cannot be viewed in isolation, instead should be evaluated in the broader clinical and radiological context (Table 2).4,6,26,27 However, unlike type 1 disease, type 2 autoimmune pancreatitis is confined to the pancreas, shows normal serum IgG4 levels, and thus neither of these parameters help in distinguishing this disease from pancreatic carcinoma. The imaging findings are similar to type 1 disease. The age of the patient is of considerable help, the majority of patients are younger than 50 years, a relatively uncommon demographic for pancreatic carcinoma. A strong association exists between type 1 autoimmune pancreatitis and inflammatory bowel disease: one-third of cases in the authors’ series were associated with inflammatory bowel disease. In summary, the two most robust pieces of corroborative evidence for the histologic diagnosis of type 2 autoimmune pancreatitis are: 1) radiologic findings, and 2) a history of inflammatory bowel disease.
Differential diagnosis Histology represents the single best gold standard for the diagnosis of type 2 autoimmune pancreatitis, and unlike type 1 autoimmune pancreatitis, there is no biomarker for this disease. In and of itself, granulocytic epithelial lesions are not diagnostic of type 2 autoimmune pancreatitis (Figure 12). Neutrophilic aggregates may be identified with diseases such as alcoholic and other forms of acute and chronic pancreatitis, groove pancreatitis, and most significantly, adjacent to a
Malignancy and autoimmune pancreatitis There is considerable uncertainty about the risk of malignancy in individuals with autoimmune pancreatitis. Bona fide cases of autoimmune pancreatitis progressing to malignancy have been documented.28,29 Low-grade PanIN is ubiquitous: in one study 82% of autoimmune pancreatitis cases showed grade 1 PanIN while 25% showed grade 2 PanIN.28 The advanced age of patients at the time of diagnosis, the relatively low incidence of autoimmune pancreatitis, as well as the lack of long-term followup in most series, precludes a thorough assessment of the risk of malignancy in autoimmune pancreatitis. A recent study identified a disproportionate increase in cancer in the first year after diagnosis of autoimmune pancreatitis, raising the intriguing possibility that autoimmune pancreatitis, in some cases, may represent a paraneoplastic phenomenon.29
Figure 12 Type 2 autoimmune pancreatitis. A needle core biopsy shows fibrosis (left) and pancreatic acinar tissue infiltrated by a neutrophilic infiltrate. An immunohistochemical stain for IgG4 is negative (not shown). Although in and of itself this appearance is not diagnostic of type 2 autoimmune pancreatitis, in conjunction with the radiologic appearance and a history of ulcerative colitis, the findings are compatible with a diagnosis of type 2 autoimmune pancreatitis.
DIAGNOSTIC HISTOPATHOLOGY --:-
Ampullary biopsy for the diagnosis of autoimmune pancreatitis Ampullary biopsies are often used to support a diagnosis of autoimmune pancreatitis: patients with autoimmune
6
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001
PANCREATIC PATHOLOGY
pancreatitis typically show 10 IgG4 positive plasma cells per high-power field.30,31 It should be emphasized that the histologic hallmarks of autoimmune pancreatitis are seldom seen in an ampullary biopsy, and hence this assay relies solely on identifying elevated numbers of IgG4 positive plasma cells, a less than optimal approach since elevated numbers of IgG4 positive plasma cells are not specific to autoimmune pancreatitis. Ampullary biopsies from individuals with type 2 autoimmune pancreatitis are generally negative for IgG4 positive plasma cells. In summary, although in some circumstances ampullary biopsies may be informative, it is best not to rely too heavily on these results.
4 Hart PA, Levy MJ, Smyrk TC, et al. Clinical profiles and outcomes in idiopathic duct-centric chronic pancreatitis (type 2 autoimmune pancreatitis): the Mayo Clinic experience. Gut 2015. gutjnl-2015309275, [Epub ahead of print]. 5 Kamisawa T, Chari ST, Lerch MM, et al. Recent advances in autoimmune pancreatitis: type 1 and type 2. Gut 2013; 62: 1373e80. 6 Okazaki K, Uchida K. Autoimmune pancreatitis: the past, present, and future. Pancreas 2015; 44: 1006e16. 7 Deshpande V, Zen Y, Chan JK, et al. Consensus statement on the pathology of IgG4-related disease. Mod Pathol 2012; 25: 1181e92. 8 Zhang L, Chari S, Smyrk TC, et al. Autoimmune pancreatitis (AIP) type 1 and type 2: an international consensus study on histopathologic diagnostic criteria. Pancreas 2011; 40: 1172e9. 9 Deshpande V, Chicano S, Finkelberg D, et al. Autoimmune pancreatitis: a systemic immune complex mediated disease. Am J Surg Pathol 2006; 30: 1537e45. 10 Deshpande V, Mino-Kenudson M, Brugge W, et al. Autoimmune pancreatitis: more than just a pancreatic disease? A contemporary review of its pathology. Arch Pathol Lab Med 2005; 129: 1148e54. 11 Mahajan VS, Mattoo H, Deshpande V, et al. IgG4-related disease. Annu Rev Pathol 2014; 9: 315e47. 12 Stone JH, Zen Y, Deshpande V. IgG4-related disease. N Engl J Med 2012; 366: 539e51. 13 Notohara K, Burgart LJ, Yadav D, et al. Idiopathic chronic pancreatitis with periductal lymphoplasmacytic infiltration: clinicopathologic features of 35 cases. Am J Surg Pathol 2003; 27: 1119e27. 14 Shinagare S, Shinagare AB, Deshpande V. Autoimmune pancreatitis: a guide for the histopathologist. Semin Diagn Pathol 2012; 29: 197e204. 15 Deshpande V, Gupta R, Sainani N, et al. Subclassification of autoimmune pancreatitis: a histologic classification with clinical significance. Am J Surg Pathol 2011; 35: 26e35. 16 Tse J, Riggi N, Ting D, et al. Diagnosis of IgG4-related disease: no longer an IS (H)-Sue? Lab Invest 2013. 184A-184A. 17 Deshpande V, Mino-Kenudson M, Brugge WR, et al. Endoscopic ultrasound guided fine needle aspiration biopsy of autoimmune pancreatitis: diagnostic criteria and pitfalls. Am J Surg Pathol 2005; 29: 1464e71. 18 Carruthers MN, Khosroshahi A, Augustin T, et al. The diagnostic utility of serum IgG4 concentrations in IgG4-related disease. Ann Rheum Dis 2015; 74: 14e8. 19 Doorenspleet ME, Hubers LM, Culver EL, et al. IgG4þ B-cell receptor clones distinguish IgG4 related disease from primary sclerosing cholangitis and biliary/pancreatic malignancies. Hepatology 2016 [Epub ahead of print]. 20 Wallace ZS, Mattoo H, Carruthers M, et al. Plasmablasts as a biomarker for IgG4-related disease, independent of serum IgG4 concentrations. Ann Rheum Dis 2015; 74: 190e5. 21 Mattoo H, Mahajan VS, Maehara T, et al. Clonal expansion of CD4 cytotoxic T lymphocytes in patients with IgG-related disease. J Allergy Clin Immunol 2016 [Epub ahead of print]. 22 Hart PA, Smyrk TC, Chari ST. Lymphoplasmacytic sclerosing pancreatitis without IgG4 tissue infiltration or serum IgG4 elevation: IgG4-related disease without IgG4. Mod Pathol 2015; 28: 238e47.
Treatment Steroids are the standard of care for the treatment of autoimmune pancreatitis. High-dose oral prednisone at a dose of 0.6e1 mg per kg per day is administered for 3e4 weeks.4,6,11 A relief of pancreatic swelling and symptoms can be observed as early as two weeks. Approximately half of all patients with type 1 autoimmune pancreatitis relapse, while those with type 2 disease seldom relapse, and recurrence with the latter disease is confined to the pancreas. There is no consensus about the duration of steroid treatment, and the benefit of maintenance steroids continues to be debated. In cases with relapse, three treatment regimens have been used: 1) high-dose steroids followed by maintenance therapy, 2) high-dose steroids without maintenance therapy, and 3) rituximab induction with or without maintenance rituximab. Azathioprine, 6-mercaptopurine, mycofenolate mofetil (MMF) have been used as steroid sparing agents for maintenance therapy. Rituximab has been shown to be particularly effective in cases that are resistant to steroid therapy and as a steroid-sparing agent.32e34
Conclusions The diagnostic approach of autoimmune pancreatitis relies on the coexistence of supportive clinical, laboratory, radiologic and histologic findings, none of which in and of themselves are pathognomonic. Nevertheless, histology represents the single best gold standard for a diagnosis of IgG4 related disease. The ultimate goal of the team of physicians is to avoid the morbidity and mortality of a pancreatic resection in individuals with autoimmune pancreatitis. Prompt additional investigations to exclude malignancy should be undertaken in individuals that lack a swift response to immunosuppressive therapy. A REFERENCES 1 Kawaguchi K, Koike M, Tsuruta K, et al. Lymphoplasmacytic sclerosing pancreatitis with cholangitis: a variant of primary sclerosing cholangitis extensively involving pancreas. Hum Pathol 1991; 22: 387e95. 2 Hamano H, Kawa S, Horiuchi A, et al. High serum IgG4 concentrations in patients with sclerosing pancreatitis. N Engl J Med 2001; 344: 732e8. 3 Shimosegawa T, Chari ST, Frulloni L, et al. International consensus diagnostic criteria for autoimmune pancreatitis: guidelines of the International Association of Pancreatology. Pancreas 2011; 40: 352e8.
DIAGNOSTIC HISTOPATHOLOGY --:-
7
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001
PANCREATIC PATHOLOGY
23 Zamboni G, Luttges J, Capelli P, et al. Histopathological features of diagnostic and clinical relevance in autoimmune pancreatitis: a study on 53 resection specimens and 9 biopsy specimens. Virchows Arch 2004; 445: 552e63. 24 Raman SP, Salaria SN, Hruban RH, et al. Groove pancreatitis: spectrum of imaging findings and radiology-pathology correlation. AJR Am J Roentgenol 2013; 201: W29e39. 25 Gupta RK, Xie BH, Patton KT, et al. Follicular pancreatitis: a distinct form of chronic pancreatitis-an additional mimic of pancreatic neoplasms. Hum Pathol 2016; 48: 154e62. 26 Ueki T, Kawamoto K, Otsuka Y, et al. Prevalence and clinicopathological features of autoimmune pancreatitis in Japanese patients with inflammatory bowel disease. Pancreas 2015; 44: 434e40. 27 Notohara K, Nishimori I, Mizuno N, et al. Clinicopathological features of type 2 autoimmune pancreatitis in japan: results of a multicenter survey. Pancreas 2015; 44: 1072e7. 28 Gupta R, Khosroshahi A, Shinagare S, et al. Does autoimmune pancreatitis increase the risk of pancreatic carcinoma?: a retrospective analysis of pancreatic resections. Pancreas 2013; 42: 506e10.
DIAGNOSTIC HISTOPATHOLOGY --:-
29 Shiokawa M, Kodama Y, Yoshimura K, et al. Risk of cancer in patients with autoimmune pancreatitis. Am J Gastroenterol 2013; 108: 610e7. 30 Kawakami H, Zen Y, Kuwatani M, et al. IgG4-related sclerosing cholangitis and autoimmune pancreatitis: histological assessment of biopsies from Vater’s ampulla and the bile duct. J Gastroenterol Hepatol 2010; 25: 1648e55. 31 Sepehr A, Mino-Kenudson M, Ogawa F, et al. IgG4þ to IgGþ plasma cells ratio of ampulla can help differentiate autoimmune pancreatitis from other “mass forming” pancreatic lesions. Am J Surg Pathol 2008; 32: 1770e9. 32 Khosroshahi A, Carruthers MN, Deshpande V, et al. Rituximab for the treatment of IgG4-related disease: lessons from 10 consecutive patients. Medicine (Baltimore) 2012; 91: 57e66. 33 Khosroshahi A, Bloch DB, Deshpande V, et al. Rituximab therapy leads to rapid decline of serum IgG4 levels and prompt clinical improvement in IgG4-related systemic disease. Arthritis Rheum 2010; 62: 1755e62. 34 Hart PA, Topazian MD, Witzig TE, et al. Treatment of relapsing autoimmune pancreatitis with immunomodulators and rituximab: the Mayo Clinic experience. Gut 2013; 62: 1607e15.
8
Ó 2016 Published by Elsevier Ltd.
Please cite this article in press as: Gupta R, Deshpande V, Current challenges in the diagnosis of autoimmune pancreatitis, Diagnostic Histopathology (2016), http://dx.doi.org/10.1016/j.mpdhp.2016.05.001